RESEARCH DOSSIER
Offshore Finance and Public Revenue
Abstract. This research dossier examines the effect of offshore finance on the level, territorial distribution, composition, collection and reliability of government revenue. It separates cross-border assets from annual taxable flows, the statistical relocation of corporate profit from legally collectible tax, host-jurisdiction fees from the revenue of source and residence countries, and gross receipts from net fiscal effects. Drawing on public finance accounting standards, published empirical research, government budget forecasts and international monitoring through 2026, it evaluates corporate base erosion, household noncompliance, tax competition, financial transparency and minimum taxation. Explicit counterfactuals and reproducible illustrative calculations show why frequently cited global tax-loss estimates, profit-shifting shares, offshore assets and cumulative enforcement yields cannot be aggregated. The synthesis identifies the fiscal conditions under which individual jurisdictions can gain while other jurisdictions lose, and explains how evidence on activity, reporting, assessment and collection supports policy evaluation. The work is a source-based comparative synthesis and does not claim a new empirical global estimate.
Keywords: Offshore finance; public revenue; fiscal capacity; corporate income tax; profit shifting; tax avoidance; offshore wealth; tax evasion; tax administration; minimum taxation; international tax cooperation; public finance
Introduction
Public revenue is the financial capacity that a government acquires through its legally and economically constituted revenue claims. Offshore finance organizes transactions, ownership and intermediation across tax jurisdictions. The fiscal importance of the offshore system arises from the potential divergence between where economic activity occurs, where income or assets are reported, where a liability is assessed, and where public receipts ultimately accrue.
A country can collect revenue from services provided to nonresidents while another country experiences an erosion of its income-tax base. An investment intermediary can avoid multiple layers of taxation while the final investor remains fully liable for residence-country tax. A change in an offshore entity's declared address can alter statistical profit allocation without altering factories, households' savings, or the stock of productive capital. These cases require different fiscal interpretations. The object of this dossier is the change in actual and potential public receipts attributable to the arrangement, including the allocation of receipts among states, the costs of maintaining fiscal capacity and the quality of the resulting tax base.
The research is situated within Offshore Economics and public economics. The neighboring studies of tax competition, BEPS, the minimum tax, tax incidence, offshore wealth and small-state development investigate mechanisms or consequences in their own right. Here they enter as determinants of a fiscal outcome. Legal liability, the legality of specific arrangements, and criminal adjudication remain matters of jurisdiction-specific law. A statistical indication of a tax-sensitive arrangement cannot establish individual wrongdoing.
The central question is how a fiscally accountable evaluation can distinguish base relocation from loss of public revenue and identify which governments gain or lose, when, and by what mechanism. The study develops this argument through a comparative synthesis of fiscal accounting standards, published empirical research, government budget documents and international monitoring. It uses transparent arithmetic illustrations to evaluate claims under alternative counterfactuals.
Public revenue and the unit of account
The classification of revenue determines the outcome being studied. The IMF's revenue classification chapter supplies a statistical distinction between transactions increasing net worth and those involving assets or liabilities. Borrowing and the repayment of principal are not revenue simply because cash flows through a government account. The Government Finance Statistics Manual 2014 (IMF, 2014) separates taxes, social contributions, grants and other revenue. Tax refunds and inappropriate collections require adjustments; an agency that collects a tax for another government does not thereby acquire that revenue as its own. The distinction between authority and collection also matters across supranational or multijurisdictional arrangements.
The OECD tax revenue indicator defines an aggregate for government as a whole relative to gross domestic product. A revenue series that includes licensing fees, penalties or commercial receipts should therefore not automatically be interpreted as a comparable corporate tax series. Three boundaries must be determined before comparing governments: the reporting entity, the accounting basis and the period. Core government budgets can differ from consolidated general government; an assessed liability differs from cash received; calendar and fiscal years may differ.
Consider a regulator that collects a fee from a financial company. Where it is payment for an identifiable service, the transaction can be revenue other than tax under the relevant accounting classification. Where the charge is effectively a compulsory unrequited levy, its classification can differ. A budget's use of the word fee does not resolve the question by itself. Economic incidence is a separate matter: an intermediary might pay the charge directly but recover it from investors through management fees.
Gross public receipts must also be distinguished from fiscal balance. A government could receive a million dollars in license payments and spend two hundred thousand dollars on additional supervision. Its gross receipt is one million; its incremental contribution to the fiscal balance, before other effects, is eight hundred thousand. The expenditure is not silently netted against the recorded revenue statistic. Fiscal policy needs both numbers because one describes the capacity to collect and the other describes the resources left for public purposes.
In international fiscal comparisons the choice of denominator is consequential. A profit-booking operation can enlarge measured GDP without producing comparable domestic wages. An unchanged tax collection then generates a lower tax-to-GDP ratio. Conversely, a contraction of the denominator mechanically raises a ratio without a rise in receipts. Real receipts per resident, receipts relative to adjusted national income, fiscal revenue composition and the tax-to-GDP ratio offer complementary perspectives. Comparability depends on specifying which adjustment is made and documenting the series used.
Jurisdictional allocation of taxable income
An economic transaction can involve a producing jurisdiction, a customer jurisdiction, the domicile of the intermediary, the residence of the beneficial recipient and the jurisdiction of the ultimate parent. These locations may overlap, but a cross-border fiscal account must preserve them as separate analytical attributes.
Source taxation links a liability to income arising within a jurisdiction under its domestic and treaty rules. Residence taxation links the taxpayer to a jurisdiction under the applicable residence rules. A host jurisdiction can tax locally incorporated entities, locally employed workers and transactions taking place within its borders. Withholding taxation collects at payment; residence taxation may give credits for qualifying foreign payments. An exempt intermediary is not necessarily an exempt final recipient.
The taxable object is determined after accounting conventions and tax rules convert a transaction into a base. Financial-account balances and gross sales are not automatically net taxable profits. A corporation's financial statement income and its taxable income can diverge owing to depreciation, loss carryforwards, exemptions, timing differences, the recognition of income and consolidated accounts. Tax rates applied to incompatible bases cannot be compared directly.
Territorial and residence-country claims can conflict. If a source country collects 10 and a residence country allows a full credit against a liability of 20, the ordinary combined result is 20 rather than 30. Reassigning the first 10 changes the territorial allocation even if the combined tax remains unchanged. If the residence tax exempts the income instead, the source charge becomes decisive. The appropriate counterfactual must therefore include the interacting systems of relief.
The location of an entity is an observable fact about registration or residence. The economic classification of an entity as an offshore intermediary depends on the functions it performs and its international relationships. The question of fiscal detriment requires a third step: tracing the base, applying applicable tax treatment and specifying the alternative arrangement. A local company may pay the locally applicable taxes and yet a different country may have a legitimate concern about an eroded base. Both facts can coexist.
Fiscal counterfactuals and revenue decomposition
Revenue lost is a counterfactual difference, not a directly observed financial transaction. Let R(j,t,a) denote general-government revenue in jurisdiction j during period t under arrangement a. Let a0 be an explicit comparison arrangement. The difference R(j,t,a0) minus R(j,t,a) measures foregone revenue under that comparison; negative values indicate an increase relative to the alternative. The arithmetic is an organizing identity, not a proposed structural economic model.
For a simple corporate base, the difference in hypothetical tax receipts can be separated into changes in the base and in the effective burden. With baseline base B0, observed base B1, baseline rate r0 and observed rate r1, the difference r0 B0 minus r1 B1 equals r0(B0 minus B1) plus B1(r0 minus r1). This identity is exact for the simplified proportional illustration. Actual tax calculations have nonlinearities, credits, deductions and loss treatment; the variables should not be confused with statutory rates or all government revenue.
Consider a hypothetical group with taxable profits of 100 in a producing country A subject to a 25% rate. Under a comparison arrangement, the company allocates 40 of those profits to an offshore jurisdiction B, where the assumed tax rate is 5%, with no other tax, relief or real response. A now collects 15 rather than 25, B collects 2 rather than zero, and the worldwide receipts are 17 rather than 25. The loss to A is 10; the gain to B is 2; the net global difference is 8. A separate license fee of 1 in B would reduce the combined gross-public-revenue difference to 7 if it were additional and did not affect other tax bases. This is a synthetic example, not an estimate for any jurisdiction.
| Hypothetical item | Baseline | Offshore arrangement | Difference, arrangement minus baseline |
|---|---|---|---|
| A corporate tax | 25 | 15 | -10 |
| B corporate tax | 0 | 2 | +2 |
| Combined corporate tax | 25 | 17 | -8 |
Now suppose 10 of the shifted 40 reflects real economic activity that would relocate even if the paper allocation were changed. The policy comparison no longer restores the entire 40 to A. Holding production fixed answers a short-run accounting question; allowing production to respond addresses the consequences of a particular policy. A universal loss number cannot be derived independently of these choices.
A serious fiscal evaluation should identify whether the comparison is full compliance under current law, elimination of a preferential regime, a different allocation rule, or a world without the intermediary. Each comparison produces a different revenue claim. Tax evasion is ordinarily benchmarked against existing liabilities. Tax avoidance may be evaluated against current law after a contested arrangement is disallowed, or against an alternative policy. Neither benchmark should be presented as the other.
Corporate profit shifting and base erosion
Multinational firms operate across jurisdictions and assign income and expenditure to separate entities. Differences in tax treatment create incentives affecting related-party prices, the allocation of finance, intangible returns and corporate structure. The economic question is the strength of the tax response and its effect on taxable income. The review by Beer and colleagues (Beer, de Mooij, and Liu, 2018) surveys several major avoidance channels and synthesizes their estimated magnitude. Its working-paper summary reports an average association corresponding to about a 1% increase in reported pretax income for a one-percentage-point reduction in the corporate tax rate. Estimates from heterogeneous studies summarize relationships across samples rather than establish a constant for every country.
The review's channel distinction is important because different mechanisms generate different fiscal traces. An intragroup payment may reduce the payer's tax base while generating an income item elsewhere. An exemption can affect a rate without moving the underlying activity. Deferral changes when a liability becomes due and may create a present-value loss even if nominal tax is eventually paid. Reorganization can alter the allocation of taxing rights without a contemporaneous change in total global pretax earnings.
Analysis by Tørsløv, Wier, and Zucman (2023), The Missing Profits of Nations, uses the exceptional profitability of foreign affiliates in selected low-tax economies to estimate shifted multinational profits. Their preferred estimate is that 36% of multinational profits were shifted in 2015. The denominator is profit earned outside the parent jurisdiction, as clarified in the 2021 prepublication manuscript. This is a study-specific estimate of profit location, not 36% of all corporate tax receipts or all corporate profits. The method depends on the credibility of its comparison of local and foreign businesses, the availability of affiliate accounts and the rules used to reallocate the estimated excess profit.
The BEPS Action 11 report (OECD, 2015) estimated world corporate income tax losses at $100–240 billion annually at 2014 levels. It treats data limitations as fundamental and uses several diagnostics rather than a single observational criterion. The 2014 estimate neither supplies a current-year fiscal loss nor needs to agree arithmetically with a later study of multinational profit shares. The denominator, counterfactual, sample and valuation date differ.
These studies establish that statistically meaningful tax-sensitive reporting warrants serious fiscal investigation. They do not eliminate uncertainty over the degree of real activity, valid contractual returns, the policy change available to a particular country, or the extent of enforceability of a proposed reform. A policymaker evaluating the incremental return from anti-avoidance enforcement needs an estimate of assessable taxable differences and the probability of collection, not only an estimate of accounting profits attributable to another location.
Private cross-border assets and household taxes
The stock of financial wealth held abroad shapes potential tax bases through interest, dividends, realized capital gains, taxable transfers or wealth charges where applicable. The stock is not an annual tax flow. A lawful diversified portfolio may be held outside the owner's residence country for currency exposure, professional management, asset allocation or other reasons and may be properly reported.
The 2018 study of offshore household wealth by Alstadsæter, Johannesen, and Zucman assembles quantitative estimates and provides an author archive of underlying research materials. Its research object is the stock of assets attributed to residents. To estimate forgone annual revenue from such wealth, it is necessary to establish taxable returns, ownership and residence, the reporting gap, applicable rates, exemptions, basis adjustments and foreign tax credits. Applying a personal tax rate directly to gross portfolio assets ordinarily confuses a stock with its return.
As a transparent illustration, suppose a resident household holds 100 units of offshore assets, of which 60 are already fully declared and 40 are not. Suppose the unreported assets generate taxable income of 3% annually and the incremental residence-country tax rate is 25%, with no foreign credit, exemption or capital loss. The illustrative income-tax gap is 40 times 0.03 times 0.25, or 0.30 units per year. This is distinct from the 40-unit unreported stock. Substituting a 1% annual wealth tax would produce a different liability under that different hypothetical regime. Past years, penalties and interest belong to separately measured categories.
Alstadsæter, Johannesen, and Zucman (2019) investigate tax evasion and inequality using information from offshore institutions matched to Scandinavian administrative records. Their estimate for the wealthiest 0.01% of households is that roughly a quarter of taxes are evaded, in contrast with much lower evasion detected in conventional random-audit data. The finding addresses a specific population and evidence design. It cannot be taken as an estimate that a quarter of taxes is evaded globally or that every wealthy household with foreign assets evades its liabilities.
A tax authority's risk-based investigation and a representative random audit answer different questions. Targeted investigations may identify highly concentrated liabilities inaccessible in conventional audit samples; their observed case-specific yield overstates average yield if extrapolated indiscriminately. Random audits provide a known sampling basis but may not detect assets that cannot be independently matched. A combination of reporting and verification can improve the observable tax base without assuming that all international investment is illicit.
Public revenue in offshore host jurisdictions
Financial specialization generates domestic fiscal receipts through mechanisms that differ from the effects on foreign tax bases. Providers can pay corporate taxes, registration and supervisory charges; employees and owners can contribute through personal taxes, consumption or transaction levies; residents and nonresidents may also purchase locally produced services. The composition depends on the host government's specific revenue regime and the scale of genuine domestic activity.
The Cayman Islands' 2026 and 2027 Plan and Estimates (Cayman Islands Government, 2025) forecasts core government revenue in 2025 of CI1,136.5 million, an upward revision of CI9.4 million in partnership fees, CI5.7 million in exempt-company fees. The partnership amount appears twice in the passage and is counted only once here.
The three distinct amounts sum to CI48.0 million aggregate upward revenue revision, this is about 48.5%. The calculation compares selected forecast deviations, not actual collected receipts or the sector's total revenue contribution. The other deviations in the budget may include positive and negative items, so the quotient must not be interpreted as a decomposition using exclusively positive weights. The text is an official budget forecast rather than an audited final account.
This example shows why the fiscal standpoint matters. An offshore host may obtain stable receipts by charging services used by an international clientele even when the amount of local taxable profit is modest. High gross offshore financial positions do not necessarily translate into a large government receipt. Conversely, fee receipts can be fiscally important in a small population even if the same amount is insignificant in the worldwide tax base.
Assessing the fiscal balance requires additional information: the incremental cost of supervision, registry administration, compliance infrastructure, public services associated with the workforce, contingent public exposure and the opportunity cost of incentives. A budget forecast alone does not establish the net benefit, nor does it identify tax that foreign governments would collect if the host's activities ceased.
The host's budget is also exposed to the elasticity of its clients' demand. Annual entity charges may be comparatively stable if legal continuity and switching costs make registrations persistent. Fee changes can still alter registrations, restructurings and the competitive position. A government considering an increase must evaluate the post-change base and its resilience, not merely multiply the current client count by a higher tariff.
Comparative fiscal structures in international financial centers
The Cayman evidence concerns a fee-rich budget and an upward forecast revision. The following cases add three different fiscal mechanisms: multinational income-tax concentration, an international financial sector embedded in the domestic earnings base, and the introduction of a new corporate tax. These cases are purposively selected to demonstrate mechanisms; their amounts retain their original currency, period and accounting basis.
Ireland's tax administration records actual payments by company ownership and separates a major exceptional receipt. In its 2024 corporation tax analysis, Irish Revenue (2025) reports EUR 28.115 billion in net corporation tax receipts excluding the proceeds of a Court of Justice of the European Union ruling. Foreign-owned multinational companies contributed EUR 24.798 billion, or around 88%; Irish-owned multinationals contributed EUR 1.206 billion, and non-multinationals EUR 2.111 billion. The ten largest corporate taxpayers accounted for 57% of the adjusted total. Ownership is defined at the level of the ultimate group and does not establish where each euro of taxable profit is economically attributable.
These data reveal a material concentration of fiscal exposure. They do not identify the share of receipts caused by profit shifting. A multinational may have both substantial productive operations and international profits recorded within its corporate structure. The appropriate decomposition requires independent evidence on activity, accounting allocation, effective taxation and the policy alternative. Separating the court-related receipts also illustrates why extraordinary collections cannot establish the long-run capacity of the tax system.
If the foreign-owned component of the adjusted 2024 account were to decline by 15%, holding all other amounts fixed, the arithmetic reduction would equal EUR 3.7197 billion, or about 13.2% of adjusted corporate receipts. This is a sensitivity analysis, not a forecast or an estimated effect of a particular international reform. It quantifies the exposure embedded in an observed fiscal composition. The state may be able to save part of unexpectedly high revenue, reduce debt, or diversify recurring revenue so that a change in the profitability or residence of a few companies does not require abrupt adjustments to services.
The Jersey 2024 annual accounts (States of Jersey, 2025) identify personal income tax of GBP 705.4 million and companies income tax of GBP 191.5 million in the financial review. The report relates growth in the former to earnings, including financial-sector earnings, and the rise in the latter chiefly to financial-sector profitability. Its discussion also explains the impact of additional credit-loss provisions on reported net income. The fiscal connection to financial activity therefore operates through both the incomes of individuals and the profitability of firms, while accounting provisions acknowledge that assessed amounts may not all be collected. The reported net income-tax figure should not be confused with the sum of the two gross components.
For Jersey, the relevant fiscal risk is joint exposure of distinct tax heads. A reduction in the number of registered structures might have a limited employment effect if those structures employ few residents. A reduction in profitability might instead substantially affect corporate receipts while employment changes more slowly. Such scenarios illustrate why the number of registrations alone cannot measure the resilience of fiscal revenue. They also show the value of comparing sector-level payroll, profit and fee information with public financial statements.
Bermuda provides an institutional transition rather than a cross-sectional ranking. The government's corporate income tax announcement confirms that its corporate income tax took effect on January 1, 2025. The 2025 Fiscal Responsibility Panel report documents the fiscal and administrative transition. At that point the original BMD 187.5 million corporate tax budget for fiscal year 2025/26 had been revised to BMD 200 million, and the medium-term projections envisaged BMD 600 million in each of the next two fiscal years. These are dated historical forecasts, not realized collections or the latest forecast. Their uncertainty reflects both the variation in multinational profits and the operation of tax credits.
The introduction of a domestic claim can change the territorial allocation of receipts even before total worldwide taxes rise. Some domestic tax may substitute for liability that another taxing jurisdiction would have assessed. An international business center can thus have strong fiscal incentives to establish a domestic corporate tax in response to international coordination. The investment required for administering that tax also imposes an immediate budgetary cost while the associated collection stream is developing.
| Jurisdiction | Measurement basis | Exposure illustrated |
|---|---|---|
| Cayman Islands | 2025 core-government forecast variances | Registration and service-related charges |
| Ireland | 2024 adjusted net corporate payments | Concentration in multinational taxpayers |
| Jersey | 2024 annual financial accounts | Profit, employment income and collection provisions |
| Bermuda | December 2025 projections | New corporate tax and credit uncertainty |
The four cases establish distinct fiscal channels, not a common legal classification as tax havens. Each jurisdiction's fiscal trajectory depends on a different combination of nonresident demand, resident economic activity, corporate taxation, regulatory costs and forecast error. Neither gross international financial positions nor a single headline tax rate can serve as a sufficient indicator of public revenue dependence.
Developing economies and fiscal capacity
The fiscal effect of international income relocation depends on both the size of the affected base and the state's ability to raise alternative revenue. Two states with the same absolute reduction in corporate receipts may face entirely different costs in terms of public services or debt financing.
The study Base Erosion, Profit Shifting and Developing Countries by Crivelli, de Mooij, and Keen (2015) examines a long international panel to distinguish spillovers through physical activity from spillovers through avoidance. The authors conclude that adverse effects on tax bases are an important concern for developing countries. This finding is not a country-specific estimate of the effect of any one offshore jurisdiction or proof that every reduction in a developing country's tax base derives from avoidance.
A fiscal comparison with later evidence is particularly useful. Garcia-Bernardo and Janský (2024), in Profit Shifting of Multinational Corporations Worldwide, estimate that more than $850 billion in corporate profits were shifted in 2017. Their country-by-country reporting approach identifies comparatively larger estimated revenue losses as a share of total tax receipts for lower-income jurisdictions. The figure measures shifted profits, not forgone tax. The authors explicitly address reporting duplication and show that country allocations depend on the functional form used for estimating shifting. This reinforces the importance of stating both the numerator and the denominator of a fiscal exposure estimate. Their evidence and the earlier IMF panel work employ different data and identification assumptions; agreement on relative fiscal vulnerability strengthens the interpretation without establishing a precise causal loss for any particular country.
Fiscal vulnerability can be amplified when a limited number of large taxpayers account for a substantial share of corporate revenue, because a legal dispute, restructuring, mining cycle or financial-sector shock can produce a material budget movement. A government with limited access to borrowing and a narrow domestic tax base has fewer means of smoothing a temporary shortfall. A state whose administrations lack comparable information or specialist personnel also faces higher marginal costs of collecting complex cross-border liabilities.
Revenue mobilization is consequently a question of institutions as well as tax design. Regulation capable of defining a liability does not guarantee that the public authority has information to identify the taxpayer, resources to audit the transaction, the ability to resolve cross-border disagreements, or channels to collect the eventual assessment. The expected revenue from a reform must incorporate these constraints.
There are also positive fiscal channels from international finance: access to investment finance, efficient pooling of capital, diversification and the facilitation of cross-border projects can enlarge real domestic production and the associated tax base. An estimation strategy that measures only direct tax payments misses these effects. Equally, attributing all new investment to a particular fiscal concession overstates its benefit when the investment would have occurred without the concession.
A public finance assessment should compare the fiscal contribution of each realistic policy alternative against its administration requirements. It can then assess the timing of revenue, uncertainty in actual collection, the proportion of yield attributable to a handful of taxpayers and the recurrent expenses supported by the incremental revenue. This approach avoids using the global size of offshore finance as a substitute for measurable development finance.
Measurement systems and identification
The principal data sources address different units of observation: government budgets measure public receipts; multinational tax reports describe group activities in jurisdictions; bank statistics measure positions and transactions; enterprise surveys record employment and assets; household evidence identifies individual ownership and reporting. Their integration requires an explicit concordance. The same asset may appear in a bank liability, a fund balance sheet and the household's securities holdings. Adding those amounts would repeat the underlying economic claim.
The OECD's 2026 country-by-country reporting chapter reports fiscal-year 2023 data supplied by 60 jurisdictions for approximately 9,400 large multinational groups. It notes that profit and activity mismatch indicators remain elevated in investment hubs. It also reports that large MNEs accounted for an average of 44.5% of total corporate tax revenues in the reporting jurisdictions. The cohort and data year matter: this is an average over reporting jurisdictions using the specified data, not the worldwide share of all public revenue. The chapter itself flags the economic turbulence surrounding FY2023 and its possible effect on these indicators.
Diagnostic ratios such as reported profits per employee compare accounting outcomes with real-activity proxies. They are useful for screening, but a financial entity may manage substantial intangible assets with a relatively small workforce. Profit rates can differ by industry, risk, investment age, losses and capital structure. Aggregate anomalies call for further investigation rather than a mechanical reallocation proportional to headcount.
Statistical profit reallocation commonly involves adjusting a comparison group or estimating the effect of cross-country tax differences. The key identification problem is whether the tax variable captures a behavioral response or correlates with omitted characteristics. Countries select tax regimes in response to investment prospects, institutions and prior patterns of activity. Firms select locations. Observational associations combine these decisions unless the method supplies a credible basis for separation.
Panel regressions can control for time-invariant differences and common shocks. They cannot automatically remove changing omitted determinants or reverse causality. Event studies can compare changes around a treaty, reporting or tax reform. Their validity depends on plausible pre-reform trends, exposure definition, anticipation and the absence or suitable modeling of contemporaneous policies. In international settings the untreated group may itself respond to the policy, violating the no-interference intuition of a simple comparison.
Administrative tax data can identify assessed and collected liabilities directly for particular taxpayers. Confidentiality restrictions, underreporting and the selection of audited cases constrain generalization. Aggregate CbCR data can evaluate allocation patterns without exposing individual tax returns but cannot establish specific liabilities. The strongest fiscal inference combines a well-identified behavioral response with a legally and administratively plausible mapping from that response to tax collections.
A comparison of estimated annual revenue losses across sources must align the numerator, domestic and worldwide denominator, reporting year, currencies, treatment of credits and alternative policy. A global loss estimate and an individual state's fiscal statistical series cannot be reconciled by adjusting exchange rates alone. The estimands differ before the data are even converted.
Information exchange and collectible revenue
Tax authorities obtain public receipts by identifying liabilities, securing assessments and collecting payments. Cross-border reporting reduces information asymmetries when the reporting institution identifies the account holder or relevant controlling party and transmits usable information to the authority with fiscal jurisdiction.
Johannesen and Zucman (2014) analyze the early G20 bank-secrecy crackdown. Their panel evidence points to the relocation of deposits to places not covered by a bilateral agreement with the depositor's country. A fall in observed deposits in one jurisdiction therefore cannot be treated as equivalent to repatriation or an increase in tax revenue. The finding concerns the particular bilateral regime studied; it should not be mechanically extrapolated to comprehensive multilateral arrangements.
Johannesen, Langetieg, Reck, Risch, and Slemrod (2020), in Taxing Hidden Wealth, examine US enforcement using administrative data. They estimate that about 50,000 individuals disclosed offshore accounts with an aggregate value of approximately $100 billion. These asset values measure the reporting response, not $100 billion in revenue. The published article abstract separately estimates an annual rise of $2–4 billion in reported capital income, corresponding to $0.6–1.2 billion in additional tax revenue. That last range is an estimated annual fiscal implication rather than a direct record of treasury cash receipts. The distinction is central: one enforcement intervention can change asset disclosure, taxable income and estimated collections by quite different magnitudes. These ranges refer to the published article rather than its earlier working-paper estimates.
The Global Forum's 2025 New Delhi announcement reports exchanges about 171 million accounts holding almost EUR 13 trillion in 2024. It also reports at least EUR 135 billion in accumulated additional identified revenue from transparency-related measures, including tax, interest and penalties, with around EUR 48 billion attributed to developing countries. The figure is cumulative and bundles distinct measures. It is not a controlled estimate of the incremental annual return to automatic exchange by itself, and it should not be added to an annual BEPS-loss estimate.
The OECD's 2025 automatic exchange peer review distinguishes legal and operational implementation. Reporting institutions, financial administrations and jurisdictions must maintain accurate due diligence, secure transmission and usable analytical processes. A formal agreement creates a route for information; effective implementation creates usable evidence; successful enforcement can create collected revenue.
The distinction matters when governments publish performance metrics. A large volume of transmitted records measures activity. The proportion matched to resident taxpayers measures usability. The value of additional liabilities assessed measures an administrative outcome. The value of confirmed collections measures receipts. Subsequent ordinary filing behavior measures a potential recurring impact. These metrics have different units and denominators and should be listed separately.
Deterrence can yield revenue even where no new penalty is imposed. Conversely, an enforcement announcement can draw forward payments that otherwise would have been received in a later period. The evaluation must consider both the level and timing of liabilities across several years. The opportunity to shift activity or change intermediaries makes network-wide coverage important, but universal claims of effectiveness remain unwarranted without evidence on the specific implementation.
The Global Tax Evasion Report 2024 (EU Tax Observatory, 2023) offers a broader assessment of the era of automatic information exchange. Its headline estimate indicates that the extent of offshore tax evasion declined roughly threefold over the preceding decade. This estimated change cannot be directly equated with an amount of tax collected. The report's wider coverage includes both legal and illegal forms of tax minimization; in this dossier its estimate serves specifically as evidence concerning offshore evasion. The early bilateral arrangements and subsequent multilateral reporting differ in geographic coverage and timing, while the American enforcement episode concerns specific administrative initiatives.
A fiscal evaluation of these policies should distinguish changes in nonreporting conditional on wealth ownership, changes in the location of assets, changes in taxable income and changes in assessed or collected liabilities. A stable aggregate of foreign assets is compatible with improved reporting. A fall in deposits held at selected centers may instead reflect reallocation. The combination of indicators, with explicit reporting of their coverage and definition, is more informative than a single measure of offshore wealth.
Tax competition and the revenue response
Tax competition gives jurisdictions incentives to adjust rates, tax bases and administrative arrangements to attract activity or taxable income. Public revenue depends on the effective burden multiplied by the base under a particular policy and on the other receipts associated with that base. An increase in the headline rate does not automatically maximize collections, because firms and investors respond through real and reporting decisions.
With a small change in a proportional hypothetical tax rate r applied to a differentiable taxable base B(r), the derivative of tax receipts is B(r) plus r times the derivative of B with respect to r. A sufficiently negative base response can offset the direct rate effect. This textbook identity does not establish the turning point for any actual country; it shows why estimating an elasticity is indispensable. It also leaves aside other public revenues, foreign responses, externalities and enforcement costs.
For a tax concession designed to attract investment, the appropriate comparison excludes projects that would have undertaken the same investment under the ordinary regime. Consider a hypothetical investment yielding taxable income of 100 annually, taxed at 25% absent a concession and at 10% under a concession. If the project would have occurred anyway, the direct difference is 15 annually before associated effects. If the project would otherwise have been located abroad, the host collects 10 rather than zero, holding other effects fixed. The same fiscal provision can therefore have a different incremental effect across recipients.
Strategic interactions make national and global outcomes diverge. A low-tax jurisdiction can gain public receipts by attracting mobile taxable income even where the redistribution reduces total receipts across countries. Conversely, coordinated rules that reduce the payoff from income relocation may raise aggregate receipts while leaving particular host governments with lower revenue from ancillary services.
A sound comparison asks whether additional private activity is genuinely induced by the fiscal measure, whether it is located in the jurisdiction for the duration of the incentive, whether it generates tax in the ordinary system afterward and whether public-service costs rise. Governments should estimate uncertainty around these effects and distinguish the expected fiscal return from the wider objective of investment policy.
Minimum taxation and changes in taxing rights
The global minimum tax offers an important illustration of the difference between raising the effective burden on a base and allocating that revenue. The OECD's Pillar Two Model Rules portal documents the jurisdictional top-up mechanism, the Income Inclusion Rule, Undertaxed Profits Rule and Qualified Domestic Minimum Top-up Tax. Their operation depends on actual domestic implementation and coordinated priority; the portal is not evidence of qualification for any particular jurisdiction. A system of top-up taxes can affect the overall liability of in-scope corporate groups where the relevant jurisdictional effective tax rate is below the applicable minimum. Which state receives receipts depends on implementing domestic rules, their interaction and the jurisdiction's position in the corporate group. Definitions, exclusions, safe harbors, accounting adjustments and transition rules affect the actual comparison.
The 2024 OECD working paper by Hugger and colleagues, The Global Minimum Tax and the Taxation of MNE Profit, evaluates expected impacts under a modeled global implementation. It identifies effects on incentives to relocate taxable profits and projected corporate receipts. Its results represent an ex ante exercise conditioned on data and modeled behavioral responses, not measured cash collections.
Contreras, Hugger, and Zawisza (2026) move toward empirical evaluation in MNE Responses to the Global Minimum Tax, using a difference-in-differences approach comparing groups around the EUR 750 million scope threshold. Such a design needs comparable trends across the threshold, attention to anticipation and changing group size, correct treatment assignment, and caution about the representativeness of groups near that boundary.
The OECD's July 2026 economic-impact update reports preliminary first-year evidence of increases in effective tax rates for in-scope businesses relative to other firms, without statistically significant declines in investment or employment over that short period. Statistical insignificance is not evidence of exact zero effects. A first-year result cannot establish the long-run equilibrium where investment, organizational structure and national policies may have adjusted.
An increase in effective taxes somewhere in a multinational group is also not a statement about the treasury that collected the top-up. Consolidated financial statements can reveal a group-level tax burden while national collection data are needed for the geographical distribution. A fiscal analysis consequently requires a reconciliation between estimated group liability, domestic assessment and national cash or accrual records.
A hypothetical low-tax jurisdiction might retain a larger portion of tax revenue if it implements an eligible domestic top-up rather than leaving another jurisdiction to exercise a taxing right. It might simultaneously experience reduced registration demand or fee collections. Fiscal evaluation should measure both changes over the same horizon and under consistent rules. This dossier makes no claim about the current effective legislation or eligibility of any particular jurisdiction; those conclusions require a dated jurisdiction-specific legal analysis.
A simplified illustration isolates the allocation question. Suppose qualifying income of 100 in country H bears a covered-tax burden of 5, and assume a minimum rate of 15%, no substance-based exclusion, timing adjustment or other complication. The additional charge is 10. If H receives that charge, its total is 15 and the parent jurisdiction P receives zero. If P receives the charge, H receives 5 and P receives 10. Worldwide tax is 15 in both cases. This is hypothetical arithmetic and not the calculation of GloBE liability for any actual taxpayer.
If part of the income were to move after reform, the assumed base would have to be adjusted. Credits, transitional rules, exemptions and accounting differences can modify the effective burden. A comparison of national tax receipts therefore needs a reconciliation between the allocation rules used in the hypothetical, measured local tax liabilities and subsequent cash or accrual records. The illustration demonstrates why a policy can redistribute tax collection even when the global burden remains unchanged.
Distribution, public services and fiscal risk
Public revenue is a means of financing collective expenditure and meeting obligations. Its distributional importance depends on the private parties that ultimately bear the burden and on the spending or other taxes that change in response. A corporate tax receipt does not imply that corporate owners alone bear its full economic cost; wages, prices and investment returns can respond.
If offshore arrangements reduce revenue in the producing or residence country, the government might borrow, reduce spending, increase another levy or seek stronger compliance. These alternatives redistribute the consequences of the initial fiscal change. A reduction in a progressive income-tax base can lead to a shift toward broader consumption taxes if those are easier to administer. Whether that makes the fiscal system more or less progressive depends on the complete tax and benefit system.
A higher administrative fee in the offshore host might be borne by absent investors through higher charges, by domestic intermediaries through lower margins or by employees through changed demand. The statutory payer and economic bearer are separate analytical categories. Estimation of the ultimate incidence requires market and behavioral evidence beyond the budget records examined here.
Timing introduces fiscal risk. Governments often commit to recurrent spending on the basis of forecast revenue. An increase arising from the settlement of old offshore tax disputes is an extraordinary receipt, not necessarily a persistent increase in the ordinary tax base. Conversely, recurring declarations of previously hidden income can produce a lasting improvement if behavior persists. Revenue management should classify extraordinary receipts and recurring resources explicitly.
Small financial centers may be especially sensitive to policy changes elsewhere because concentrated sources of receipts depend on nonresident demand. Larger source and residence countries may have diversified bases but substantial exposure to individual highly profitable corporate groups. Both settings warrant scenario analysis: changes in client counts, profits, returns, exchange rates, compliance costs and regulatory requirements can produce simultaneous effects.
Finally, a fiscal risk analysis should distinguish the annual direct expense of oversight from contingent exposure. A guarantee or potential public support operation is not a current recorded revenue loss. Its expected cost can affect a broader fiscal sustainability assessment, while its realized fiscal impact belongs to the period and category specified by applicable accounting conventions.
Comparative evaluation and evidence reconciliation
The evidence assembled in this dossier concerns distinct objects. It is useful to present the numerical findings side by side precisely because they are not additive.
| Evidence | Quantity | Period and unit | Interpretation |
|---|---|---|---|
| OECD Action 11 (2015) | $100–240 billion | Annual, at 2014 levels | Modeled global CIT revenue loss |
| Tørsløv et al. (2023) | 36% | 2015, foreign profits of MNEs | Estimated fraction shifted |
| Alstadsæter et al. (2019) | About 25% | Taxes of richest 0.01% in study population | Estimated evasion, not population-wide |
| Johannesen et al. (2020) | About $100 billion | Aggregate asset value disclosed | Reporting response, not tax receipts |
| OECD CbCR (2026) | 44.5% | FY2023 average reporting jurisdictions | Covered large-MNE share of CIT revenue |
| Global Forum (2025) | At least EUR 135 billion | Cumulative, tax plus interest and penalties | Additional revenue identified across several measures |
| Cayman Islands budget (2025) | CI$48.0 million | 2025 forecast revision | Difference from original core-government revenue budget |
The full citations for these quantities occur in the corresponding chapters and in References. The table gives the minimum context required to avoid comparing incompatible observations. No sum or pooled average is meaningful across its rows. In particular, the 36% shifting share cannot be applied without qualification to the 44.5% large-MNE revenue share. The product would combine a reported-profit adjustment with a fiscal revenue fraction drawn from an unrelated sample and period.
When an authority proposes a new fiscal intervention, the relevant assessment can be organized as an evidence sequence. Identify the government and taxable object. Estimate the baseline and the attainable alternative using the same period. Establish the extent to which the proposed rules create a collectible liability. Evaluate how taxpayers and other governments respond. Reconcile gross revenue, expenditure, extraordinary amounts, credits, refunds, and cash timing. Document which results are attributable to the intervention and which reflect macroeconomic change.
A practical revenue-monitoring account can keep the following categories separately: ordinary tax receipts, government fees, extraordinary collections, refunded amounts, incremental administrative expenses, and fiscal risks that have not yet materialized. Geographic attribution should track the country exercising taxing authority rather than assuming the location of the receiving bank account determines the ultimate government. Policy evaluations should publish uncertainty ranges where data support them, rather than deriving an unsupported precise amount from an assumption about unobservable offshore activity.
The standard for moving from diagnostic evidence to a fiscal estimate is consistency between the behavior observed and the liability measured. Accounting ratios and deposit movements can inform risk assessment. An estimated change in reportable profits can inform a modeled tax base. A court or administrative determination can establish assessed liability in a defined case. Cash records establish receipts. Each stage deserves scrutiny at its own level.
Research limitations and reproducibility
This is a comparative research synthesis, not a newly estimated cross-country revenue-loss dataset. Sources were selected for their relevance to fiscal concepts, profit-location research, household reporting and enforcement, published tax monitoring, host-government budget planning and minimum-tax evaluation. The resulting source set is purposive rather than a systematic-review census of the literature.
The research distinguishes official documents, extracted passages, direct report chapters and bibliographic abstracts. Full-text access to some articles and IMF documents was restricted, so claims from those works are limited to the material retrieved. The Irish Revenue ownership table and the prepublication profit-shifting manuscript's introduction were visually checked against the PDFs. Screenshot attempts for other official PDFs did not succeed; their textual passages are identified in the evidence ledger. The Cayman and Bermuda observations concern dated projections, Ireland an adjusted payment series, and Jersey financial accounts. No underlying confidential microdata or replication packages were rerun.
The illustrative fiscal accounts in this dossier use declared units and deliberately simplified assumptions. The combined corporate-tax example follows directly from 60 times 0.25 plus 40 times 0.05; the household example follows from 40 times 0.03 times 0.25. The Cayman budget's selected variance calculation uses 9.4 plus 8.2 plus 5.7, totaling 23.3; the denominator is 48.0. These calculations can be reproduced independently without access to nonpublic information. They are pedagogical comparisons rather than empirical estimates beyond the stated budget amounts.
Useful extensions would require the complete final government accounts for comparable jurisdictions, consistent tax and fee classifications, disaggregated data on revenue attribution and administration expenditure, and an empirical strategy for separating real relocation from taxable-income reassignment. For investigations of transparency reforms, matched account and return data and independently measured collections would strengthen claims about persistence and attribution.
The limitations are substantively consequential: a single country ranking of net gains and losses is not established, nor is a worldwide annual figure combining every revenue channel. The research supports the analytical distinctions, identifies mechanisms and interprets published results within their samples. It leaves numerical answers beyond those evidence boundaries open.
Conclusion
Offshore finance affects public revenue through the relationship between taxable income, financial ownership, jurisdictional authority, compliance and collection. Its fiscal consequences are distributed across governments with different functions, revenue systems and capacities. They cannot be inferred from the gross size of cross-border assets or the fact of offshore registration.
The research establishes several conclusions. First, base relocation and revenue loss are different quantities. The conversion requires a comparison arrangement, the application of rates and bases, allowances for interacting tax systems and, where relevant, a model of behavioral responses. Second, a jurisdiction can collect fees and other public receipts from offshore financial specialization while another jurisdiction loses access to part of its taxable base. The sum of those separate consequences must be assessed on a consistent basis.
Third, household wealth, disclosed assets, assessed liabilities and realized fiscal receipts have separate dimensions. Empirical evidence of concentrated offshore noncompliance and responses to information reforms supports policies that improve tax observability and administration, while limiting extrapolation beyond the research populations and episodes actually studied. Fourth, corporate reporting data and international estimates provide important evidence of tax-sensitive profit allocation but cannot determine individual tax obligations without additional facts and applicable law.
Fifth, the reliability and territorial distribution of fiscal receipts is as relevant as their aggregate level. Host-country fees, source-country income taxes and residence-country liabilities can move in different directions during reforms. Public officials need an account that identifies the recipient, the time of realization, the recurring or extraordinary character of the receipt and the cost of securing it.
Finally, credible fiscal evaluation requires a disciplined comparison of actual and attainable alternatives. International tax coordination, transparency, enforcement and investment incentives should be examined in terms of their incremental effects on public receipts, administrative capacity, productive activity and fiscal risk. The analytical contribution of this dossier is the integration of these distinctions into a consistent reading of existing evidence. The result supports transparent, jurisdiction-specific fiscal evaluation without asserting a new named theory or a numerical conclusion that the available sources cannot sustain.
References
- Alstadsæter, A., Johannesen, N., & Zucman, G. (2018). Who Owns the Wealth in Tax Havens? Macro Evidence and Implications for Global Inequality. Journal of Public Economics, 162, 89–100. Author research archive with data and programs. Source record
- Alstadsæter, A., Johannesen, N., & Zucman, G. (2019). Tax Evasion and Inequality. American Economic Review, 109(6), 2073–2103. DOI: 10.1257/aer.20172043. Source record
- Beer, S., de Mooij, R. A., & Liu, L. (2018). International Corporate Tax Avoidance: A Review of the Channels, Magnitudes, and Blind Spots. IMF Working Paper 18/168. DOI: 10.5089/9781484363997.001. Source record
- Bermuda Fiscal Responsibility Panel. (2025). 2025 Fiscal Responsibility Panel Report. December 2025, paragraphs 25–27. Source record
- Cayman Islands Government. (2025). 2026 and 2027 Plan and Estimates. Forecast Operating Revenues, PDF page 33. George Town: Cayman Islands Government. Source record
- Contreras, L., Hugger, F., & Zawisza, T. (2026). MNE Responses to the Global Minimum Tax. OECD Taxation Working Papers 77. DOI: 10.1787/0bc9aac9-en. Source record
- Crivelli, E., de Mooij, R. A., & Keen, M. (2015). Base Erosion, Profit Shifting and Developing Countries. IMF Working Paper 15/118. DOI: 10.5089/9781513563831.001. Source record
- EU Tax Observatory. (2023). Global Tax Evasion Report 2024. Paris: EU Tax Observatory. Source record
- Garcia-Bernardo, J., & Janský, P. (2024). Profit Shifting of Multinational Corporations Worldwide. World Development, 177, 106527. DOI: 10.1016/j.worlddev.2023.106527. Source record
- Global Forum on Transparency and Exchange of Information for Tax Purposes. (2025). International Community Meets in New Delhi to Reflect on Latest Advances in Global Tax Transparency. OECD, December 2, 2025. Source record
- Government of Bermuda, Ministry of Finance. (2026). Bermuda Corporate Income Tax. Updated July 14, 2026. Source record
- Hugger, F., González Cabral, A. C., Bucci, M., Gesualdo, M., & O’Reilly, P. (2024). The Global Minimum Tax and the Taxation of MNE Profit. OECD Taxation Working Papers 68. DOI: 10.1787/9a815d6b-en. Source record
- IMF. (2014). Government Finance Statistics Manual 2014. Washington, DC: International Monetary Fund. Source record
- Irish Revenue. (2025). Corporation Tax—2024 Payments and 2023 Returns. Dublin: Revenue, Tables 1, 2 and 5, pp. 4–11. Source record
- Johannesen, N., & Zucman, G. (2014). The End of Bank Secrecy? An Evaluation of the G20 Tax Haven Crackdown. American Economic Journal: Economic Policy, 6(1), 65–91. DOI: 10.1257/pol.6.1.65. Source record
- Johannesen, N., Langetieg, P., Reck, D., Risch, M., & Slemrod, J. (2020). Taxing Hidden Wealth: The Consequences of US Enforcement Initiatives on Evasive Foreign Accounts. American Economic Journal: Economic Policy, 12(3), 312–346. DOI: 10.1257/pol.20180410. Source record
- OECD. (2015). Measuring and Monitoring BEPS, Action 11—2015 Final Report. OECD/G20 Base Erosion and Profit Shifting Project. Paris: OECD Publishing. DOI: 10.1787/9789264241343-en. Source record
- OECD. (2025). Peer Review of the Automatic Exchange of Financial Account Information 2025 Update. Paris: OECD Publishing. DOI: 10.1787/bbf150e4-en. Executive summary. Source record
- OECD. (2026). Corporate Tax Statistics 2026. Paris: OECD Publishing. DOI: 10.1787/73af6222-en. Chapter 7, Country-by-country reporting statistics. Source record
- OECD. (2026). OECD Publishes New Analysis on the Economic Impacts of the Global Minimum Tax. July 15, 2026. Paris: OECD. Source record
- OECD. (n.d.). Global Anti-Base Erosion Model Rules (Pillar Two). Paris: OECD. Model Rules published December 20, 2021; portal accessed November 19, 2026. Source record
- OECD. (n.d.). Tax revenue. OECD Data indicators. Paris: OECD. Accessed November 18, 2026. Source record
- States of Jersey. (2025). States of Jersey Group 2024 Annual Report and Accounts. Financial Review, p. 56. Source record
- Tørsløv, T., Wier, L., & Zucman, G. (2021, August 10). The Missing Profits of Nations. Prepublication manuscript. Source record
- Tørsløv, T., Wier, L., & Zucman, G. (2023). The Missing Profits of Nations. Review of Economic Studies, 90(3), 1499–1534. DOI: 10.1093/restud/rdac049. Source record